Twin Tracers, Different Signatures: Bone Scans for Cardiac Amyloidosis Are Not Interchangeable

Two radioactive bone-scanning agents long treated as interchangeable twins in the diagnosis of cardiac amyloidosis in fact behave in strikingly different ways once they enter the body, according to a retrospective multicenter study of 224 patients conducted at two Italian referral centers in Bologna and Florence and published in the European Journal of Nuclear Medicine and Molecular Imaging. The work compared technetium-99m labeled DPD, used in Bologna, with technetium-99m labeled HMDP, used in Florence, in patients investigated for suspected cardiac amyloidosis between 2015 and 2021. Both agents are cornerstones of the non-invasive diagnosis of transthyretin cardiac amyloidosis, a progressive disease in which misfolded transthyretin protein deposits as amyloid fibrils in the heart muscle, stiffening the ventricular walls and eventually causing heart failure. Yet the new analysis shows that each tracer paints a distinct whole-body picture, and that ignoring those differences can distort how physicians grade disease severity on a scan.

The research team enrolled patients whose final diagnoses had been confirmed as wild-type transthyretin amyloidosis, hereditary transthyretin amyloidosis, or light-chain AL amyloidosis, the latter driven by abnormal antibody light chains rather than transthyretin. The Bologna cohort comprised 109 patients imaged with DPD, including 70 with wild-type ATTR, 20 with hereditary ATTR, and 19 with AL disease. The Florence cohort comprised 115 patients imaged with HMDP, including 90 with wild-type ATTR, 13 with hereditary ATTR, and 12 with AL disease. Overall, wild-type ATTR accounted for 71 percent of the population, hereditary ATTR for 15 percent, and AL amyloidosis for 14 percent. Both cohorts were strongly male predominant, at roughly 83 percent, and the two centers’ cases were read by teams of three experienced nuclear medicine physicians who underwent joint calibration before the study and were blinded to each patient’s final diagnosis.

Every participant underwent late planar whole-body scintigraphy approximately three hours after intravenous injection of 740 megabecquerels of the respective radiopharmaceutical, following current procedural guidelines. Patients were asked to hydrate well and empty their bladder before scanning to reduce background activity in the pelvis. Imaging used dual-head SPECT/CT gamma cameras with low-energy high-resolution collimators, and dedicated SPECT/CT acquisitions were performed where indicated to confirm the precise cardiac location of uptake and to anatomically characterize any extracardiac findings. The myocardial signal was graded visually with the Perugini score, a four-point scale in which grade 0 means no cardiac activity, grade 1 means uptake weaker than bone, grade 2 means cardiac uptake equal to bone signal, and grade 3 means intense myocardial retention with minimal or absent bone uptake. Beyond the heart, the readers systematically recorded tracer retention in the lungs, the limb musculature, the abdominal viscera, and miscellaneous sites such as the tongue and salivary glands.

The most eye-catching divergence emerged in the visual grading of wild-type ATTR patients. In the DPD cohort, 76 percent of wild-type ATTR scans showed Perugini grade 2, while only 21 percent reached grade 3. In the Florence HMDP cohort the pattern was essentially inverted: 72 percent of wild-type ATTR patients scored grade 3 and just 19 percent scored grade 2, a difference that was highly statistically significant. The authors attribute this skew not to any real difference in disease severity between the two patient groups but to the distinctive whole-body kinetics of the two molecules. DPD has a high affinity for amyloid deposits lodged in the skeletal muscles of the limbs and shoulder and hip girdles. That diffuse soft-tissue retention acts as a competitive background, a so-called soft-tissue sink effect, which dampens the heart-to-body contrast and frequently pins the visual score at grade 2 even when the disease is advanced. HMDP, by contrast, clears rapidly from soft tissues, raising the target-to-background ratio and visually amplifying the myocardial signal, which predisposes scans to a grade 3 classification.

The extracardiac biodistribution data reinforce that interpretation. Among wild-type ATTR patients, appendicular muscle uptake appeared in 84 percent of DPD scans but only 38 percent of HMDP scans, a difference with a p-value below 0.001. A similar trend appeared in the hereditary ATTR subgroup, where muscle uptake was present in 80 percent of DPD scans versus 38.5 percent of HMDP scans. HMDP showed the mirror-image behavior in the lungs: pulmonary tracer retention was observed in 68 percent of wild-type ATTR patients imaged with HMDP versus 46 percent imaged with DPD, and in 61 percent of hereditary ATTR patients imaged with HMDP versus a mere 10 percent imaged with DPD. This distinctive pulmonary signature of HMDP, quantified with the established Janssen Score, matters for daily reporting because unfamiliar readers might otherwise mistake lung retention for a technical artifact or for unrelated interstitial lung disease. Miscellaneous uptake, in sites such as the tongue and salivary glands, was also more frequent with DPD in the wild-type ATTR group, at 89 percent versus 74 percent.

Not every scintigraphic feature depended on the tracer’s chemistry. Visceral uptake, meaning tracer retention in the liver or spleen, was rare in both ATTR subtypes across both centers and showed no significant difference between the two agents. Where it did appear, it was overwhelmingly a hallmark of AL amyloidosis, occurring significantly more often in light-chain patients than in either ATTR group regardless of the tracer used. In multivariate logistic regression, visceral uptake emerged as a powerful negative predictor of ATTR disease, with an odds ratio of 0.016, meaning its presence strongly favored an AL diagnosis. The authors emphasize that this reflects the intrinsic biology of the fibrils rather than radiotracer pharmacokinetics: light-chain amyloid frequently deposits in abdominal organs, and both DPD and HMDP reliably flag that pattern. Neither tracer showed significant myocardial uptake in most AL patients, preserving a robust ability to exclude light-chain cardiac involvement.

The regression models also quantified which imaging findings independently predict an ATTR diagnosis over AL. Across the whole population, a Perugini score of grade 2 or higher carried an odds ratio of 30.49 for ATTR, while miscellaneous uptake carried an odds ratio of 6.35. Muscle uptake, despite its visual prominence on DPD scans, did not reach statistical significance as an independent predictor in the multivariate model, with an odds ratio of 1.32 and a p-value of 0.694, likely because it overlaps with other features. When the cohorts were stratified by tracer, the visual score was significant in both but with a larger odds ratio in the HMDP model, 55.85, than in the DPD model, 15.72, consistent with HMDP’s superior myocardial contrast. Visceral uptake remained a significant negative predictor in both cohorts. Notably, none of the scintigraphic parameters significantly distinguished hereditary ATTR from wild-type ATTR, with wide confidence intervals crossing the line of no effect, a result the authors attribute to the tracers operating at the limits of their pharmacokinetic sensitivity and to differing referral pathways between the two subtypes.

The study’s authors, led by Andrea Di Giorgio and Rachele Bonfiglioli with collaborators including Francesco Cappelli and Simone Longhi, argue that these findings should change how scans are reported. With DPD, skeletal muscle uptake should be actively assessed as a confirmatory sign of systemic amyloid burden, and clinicians should recognize that the agent tends to stabilize the visual score at grade 2, guarding against the misreading of advanced disease as early-stage involvement. With HMDP, pulmonary retention should be recognized as an expected feature rather than a lung pathology, and its absence of muscle signal should not be taken to exclude extracardiac involvement. The team also highlights the practical importance of SPECT/CT fusion imaging, which anatomically localizes the myocardial signal and separates it from overlying soft-tissue activity, resolving ambiguities that planar images alone cannot, particularly when the DPD sink effect suppresses apparent cardiac intensity.

The authors are careful to frame these patterns as probabilistic indicators that complement, rather than replace, the mandatory diagnostic workflow. A positive bone scan, even with a high Perugini grade and characteristic extracardiac uptake, must still be interpreted within the broader clinical picture, and definitive diagnosis requires the biochemical exclusion of monoclonal gammopathy and confirmatory genetic sequencing of the transthyretin gene. Scintigraphic signatures may also appear incidentally on bone scans performed for unrelated orthopedic, oncological, or rheumatological reasons, or in asymptomatic genetic carriers without overt cardiomyopathy. The study carries limitations inherent to its retrospective design, including potential selection bias, the use of different gamma camera models at the two centers without formal phantom-based cross-validation, and a patient population dominated by the Ile68Leu transthyretin variant, which may limit generalizability to other mutations even as it reduces genetic confounding.

The bottom line for clinical practice is that the two tracers remain gold standards for non-invasive identification of transthyretin cardiac amyloidosis, but they are distinct molecular entities whose readings demand tracer-specific expertise. HMDP maximizes the heart-to-bone visual contrast and often drives grades toward 3, while DPD offers a more comprehensive view of the systemic muscular amyloid burden at the cost of a compressed visual scale. Both preserved strong specificity in excluding AL disease, with hepatic and splenic uptake serving as a tracer-independent red flag for light-chain disease. As bone scintigraphy continues to anchor the diagnostic pathway for a disease increasingly targeted by new therapies, the message of this multicenter comparison is that the choice of radiotracer is not a technical detail but a variable that shapes every grade and every extracardiac finding a physician reports.

Subject of Research: Comparative biodistribution of 99mTc-DPD and 99mTc-HMDP radiotracers in scintigraphic diagnosis of cardiac amyloidosis

Article Title: The role of extracardiac radiotracer biodistribution in the evaluation of cardiac amyloidosis: a retrospective multicenter study on [99mTc]Tc-DPD and [99mTc]Tc-HMDP

Article References: Di Giorgio, A., Battisti, G., Bonazzi, N., Prisco, S. E., Starace, M., Longhi, S., Fanti, S., Berti, V., Cappelli, F., & Bonfiglioli, R. (2026). The role of extracardiac radiotracer biodistribution in the evaluation of cardiac amyloidosis: a retrospective multicenter study on [99mTc]Tc-DPD and [99mTc]Tc-HMDP. European Journal of Nuclear Medicine and Molecular Imaging. https://doi.org/10.1007/s00259-026-08153-1

Image Credits: AI Generated

DOI: 10.1007/s00259-026-08153-1

Keywords: cardiac amyloidosis, transthyretin amyloidosis, bone scintigraphy, DPD, HMDP, radiotracer biodistribution, Perugini score, SPECT/CT, AL amyloidosis, extracardiac uptake, nuclear medicine, multicenter study